Release Time:2019-03-11  Hits:

Indexed by: Conference Paper

Date of Publication: 2012-01-01

Included Journals: Scopus、CPCI-S

Volume: 1479

Issue: 1

Page Number: 2110-2113

Key Words: Bounded-but-unknown uncertainty; Non-probabilistic reliability; Probability; Optimization

Abstract: When the amount of information available on uncertain parameters is not enough to accurately define the probability distribution functions and only bounds of the uncertain parameters are available, non-probabilistic reliability are recently used. Interval variables and convex model are usually used to quantify the bounded-but-unknown uncertainty and the corresponding models of non-probabilistic reliability measure and design optimization are brought forward. Furthermore, probabilistic reliability theory can also be utilized by assuming the bounded-but-unknown variables as uniform random variables based on the principle of maximum entropy. In this paper, these three models of design optimization with bounded-but-unknown uncertainty are discussed and compared. It is pointed out that non-probabilistic interval model is too conservative and the probabilistic model is a rational alternative.